a systems approach for the development of a sustainable...
TRANSCRIPT
A systems approach for the development of a sustainable
community—the application of the sensitivity model (SM)
Shih-Liang Chana,*, Shu-Li Huangb
aDepartment of Real Estate and Built Environment, National Taipei University, Taipei, TaiwanbGraduate Institute of Urban Planning, National Taipei University, Taipei, Taiwan
Received 8 July 2003; revised 20 March 2004; accepted 7 April 2004
Abstract
Corresponding to the concept of ‘Think globally, act locally and plan regionally’ of sustainable development, this paper discusses the
approach of planning a sustainable community in terms of systems thinking. We apply a systems tool, the sensitivity model (SM), to build a
model of the development of the community of Ping-Ding, located adjacent to the Yang-Ming-Shan National Park, Taiwan. The major issue
in the development of Ping-Ding is the conflict between environmental conservation and the development of a local tourism industry. With
the involvement of local residents, planners, and interest groups, a system model of 26 variables was defined to identify characteristics of
Ping-Ding through pattern recognition. Two scenarios concerning the sustainable development of Ping-Ding are simulated with interlinked
feedbacks from variables. The results of the analysis indicate that the development of Ping-Ding would be better served by the planning of
agriculture and the tourism industry. The advantages and shortfalls of applying SM in the current planning environment of Taiwan are also
discussed to conclude this paper.
q 2004 Elsevier Ltd. All rights reserved.
Keywords: Bio-cybernetics; Systems thinking; Sensitivity model; Semi-quantitative simulation; Sustainable community
1. Introduction
The concept of sustainability and sustainable development
has been discussed for years since the oft-quoted World
Commission on Environment and Development (WCED,
1987) in 1987. It has been touted as a new planning agenda,
though the viewpoints regarding the meaning of sustainability
are still diverse (Beatley and Manning, 1997). Two questions
concerning the sustainability movement have frequently been
addressed. The first is the appropriate geographical scale for
action. Since ‘local action’ (a bottom-up perspective) is the
consensus approach to practical action, and since a community
can serve as the fundamental element of a hierarchical structure
of an urban area, it would be appropriate to address
sustainability at the scale of community development. In
addition to this geographical scale question, there is concern for
finding an effective method to plan and manage local
development in a sustainable manner. Much effort has been
made to develop sustainability indicators for local development
(see Huang et al., 1998). However, sustainability indicators by
themselves only partially improve our understanding of
community development. Systems thinking and an integrated
approach would be more appropriate for dealing with the
sustainability of local development, and this has become a key
focus of a number of studies (Huang and Chen, 1999; Rothman,
Robinson and Biggs, 2002; Rotmans and Asselt, 2002).
In this paper, we apply a sensitivity model (SM),1 a
systems approach developed by Vester and Hesler (1982),
as a planning tool for dealing with development issues in the
small village of Ping-Ding. With its foundations in systems
thinking and bio-cybernetic rules, SM provides us with a
convenient tool to assess the sustainability of a local
community by identifying the pertinent characteristics of a
community, and by simulating its development in scenarios
using semi-quantitative data.
Following this introduction, Section 2 will elaborate the
fundamental concepts of sustainable communities and
0301-4797/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvman.2004.04.003
Journal of Environmental Management 72 (2004) 133–147
www.elsevier.com/locate/jenvman
* Corresponding author. Address: Department of Real Estate and Built
Environment, National Taipei University, 67, Sec. 3, Ming-Shen E. Rd.
Taipei, Taiwan, ROC. Tel.: þ886-2-25009156; fax: þ886-2-2507-4266.
E-mail address: [email protected] (S.-L. Chan). 1 Internet: http://www.sensitivity-model.com.
systems thinking. The bio-cybernetic based systematic
planning tool, SM, will be described in the third section.
The SM is then applied to the case of the Ping-Ding
community in Taipei city to identify important system
characteristics in Section 4. A simulation of Ping-Ding with
some partial scenarios and discussion of results are
presented in Section 5. The last section discusses the
advantages and shortfalls of SM to conclude the paper.
2. Systems thinking and sustainable community
Sustainable development has become an influential and
widely used term, even though it has diverse meanings. The
1987 report, Our Common Future, by the United Nations
World Commission on Environment and Development
(WCED), sets forth the most widely used definition of the
concept, ‘Sustainable development is the development that
meets the needs of the present generation without
compromising the ability of future generations to meet
their own needs’ (WCED, 1987). At the Earth Summit in
Rio, considerable attention was devoted to sustainability,
and the concept was embodied in the resulting UN
Framework Convention on Sustainable Development. In
addition, the OECD, the LYNCTAD, and the US Pre-
sidential Council on Sustainable Development, and many
international policy-oriented institutions, are devoting time
and energy to the analysis of sustainable policies.
Though the concept of sustainability has been adopted in
principle, there are still some issues that seem unclear from
the practical viewpoint. The first is the issue of appropriate
geographical scale. It is argued that sustainability planning
would be less efficient in a large region (Kildow, 1992).
Under the concept of ‘Think globally, act locally and plan
regionally (Forman, 1995),’ the scale of a community,
which contains some important elements for local develop-
ment, is suggested to be a proper one. The second issue in
the implementation of sustainable development is the need
for an effective tool to assist in achieving the goal. Since a
community is a complex system of humans and natural
environment, it is necessary to deal with it from a
comprehensive and systematic viewpoint.
The sustainable development of a location should avoid
system overshoot by imposing negative feedback to system
growth when approaching carrying capacity. That is, from
the viewpoint of sustainability, a sound growth pattern of a
community should be a logistic S-curve type (see Fig. 1) to
avoid the collapse of over-development, rather than an
exponential curve that most human societies have shown
after the industrial revolution (Vester, 1988).
The definition of sustainable community varies accord-
ing to the interest, needs, and culture of different commu-
nities, but most focus on economic, environmental, and
social issues. The analogy of a three-legged stool has been
used to stress the importance of addressing and balancing
the three categories of issues, in which the legs of the stool
represent economic, social, and environmental components
(Lachman, 1997). To stand well the stool requires balance
among these three legs that are usually in conflict and
competition. In the way toward balancing the three
components, many emphasize the involvement of open
processes in which all members of a community are
encouraged to participate. The focus is on consensus
building through communication and cooperation among
Fig. 1. Logistic S-curve of sustainable development.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147134
many different interest groups from both the community and
those outside the geographic neighborhood. Such sustain-
ability activities enhance the individuals’ and organizations’
feelings of attachment, value, and connection to the
community. By means of public involvement, it is possible
to foster a sense of community, which is a critical element of
a sustainable community, in addition to a sound physical
environment.
Achieving a healthy, sustainable community requires a
long-term, integrated, and systems approach to addressing
economic, environmental, and social issues. There are two
strategies for analyzing a community system, namely the
micro and macro viewpoints. The former focuses on the
individual features of a community such as population
growth, industry or transportation issues. It argues that
progress towards sustainable development should be based
on indicators of standardized measurement (Hodge and
Hardi, 1997). Many efforts have been devoted to developing
the principles and indicators for a sustainable community
(Maureen, 1999; Huang et al., 1998; Zachary, 1995). The
macro view regards the community as a holistic system.
Since the elements of a community are dynamically
interlinked, it is likely that a community will be mis-
interpreted by overemphasizing an individual sector without
taking the interaction into account, like the old Chinese
idiom ‘seeing a tree without seeing the forest,’ or the
disconnected community as shown in Fig. 2.
A systematic approach emphasizes identifying and
describing the interactions between system components.
As shown in Fig. 2, planners frequently ignore the
interactions between land use, resource, culture, tourism
and communal life of a local community. A systematic
approach has to connect the broken linkages among the
sectors of the community, as well as define the sectors of
the community. A feedback system then is built with all the
system components and linkages. Through the linkages,
changes in one component will induce changes in another
component, which may in turn induce a change in the third
component. Many such interactions can be linked together
in chains of cause and effect relationships and chains of
cause and effect relationships can intersect themselves. This
means that a component can start a sequence of causes and
effects that eventually loops back, so that each of the
components in the loop indirectly influences itself. A system
can contain more than one feedback loop. The behavior of a
given component, in such a set of relationships, is the
outcome of multiple competing factors (Clayton and
Radcliffe, 1996).
The discussion above gives us a fundamental idea of
what is essential for managing the sustainable development
of a community. It is not surprising that there have been a
number of recent studies emphasizing public participation
and systems thinking. For example, the integrated assess-
ment (IA), in which communication between different
participants is at the very heart, has become a rapidly
evolving field in the past decade (Rotman and van Asselt,
2002). In addition, with the same philosophy of linking
indicators and providing visions that make sense to the
participants, an approach similar to SM can be found in the
QUEST project at the University of British Columbia
(Rothman, Robinson and Biggs, 2002). Vester (1988, 1999)
provides the rules of bio-cybernetics as a guideline for
Fig. 2. Torn network of local development.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 135
further application. These principles basically incorporate
the concept of feedback loops to check and balance the
performance of systems for symbiotic relationships between
humans and environment. Eight rules to evaluate the
sustainability of a community system are shown in Fig. 3.
They include: (1) negative feedbacks must dominate
positive feedbacks; (2) the vitality of a system must be
independent from quantitative growth; (3) the system must
be function-oriented and not product-oriented; (4) use of
existing forces (principle of jiu-jitsu) instead of fighting
(boxing-method); (5) multiple usage of products, functions
and organizational structures; (6) recycling processes for the
utilization of waste and sewage; (7) symbiotic reciprocal
usage of difference by employing coupling and exchange;
and (8) biological design of products, procedures and forms
of organizations by feedback planning. These rules have
been further applied to develop a systems tool for planning,
which will be introduced later in this paper.
In Section 3, the tool for implementing the systems
approach, sensitivity model, will be introduced and applied
to the Ping-Ding community. The experience gained from
the application of this SM tool can provide better planning
for sustainable development in the local context.
3. A systems approach—sensitivity model
The systems tool—sensitivity model—was developed by
Vester and Hesler in 1975 (Vester and Hesler, 1982), during
work on a UNESCO-program, Man and the Biosphere
(MAB II), to solve increasingly complex problems in the
world. It is a systems tool for planning based on Bio-
cybernetics that was intended to ameliorate dissatisfaction
with conventional planning aids. Since the development of
SM, it has been applied to many different fields of research
including corporate strategic planning, technology assess-
ment, developmental aid projects, examination of economic
sectors, city, regional and environmental planning, traffic
planning, insurance and risk management, and financial
services.
The philosophy behind SM includes many principles that
are emphasized in the literature of General System Theory.
The fundamental ideas of SM, which make it different from
other planning approaches, include systems thinking,
fuzziness, and simulation of semi-quantitative data. It
emphasizes pattern recognition and the theory of feedback
mechanisms rather than mono-causal form of recognition,
and it makes the analysis of complex systems possible by
using the approach of fuzzy logic. The application of semi-
quantitative simulation has been discussed in ecological
modeling in recent years (Jorgensen, 1997; Ecological
Modeling 1996, Vol. 85). The planners are able to capture
the examined system and its socio-economic-ecological
environment as a bio-cybernetic entity without getting lost
in a countless number of factors and variables (Vester,
1988). Most traditional quantitative simulation models
require precise parameters that might be unavailable
because of the complexity of system relationships. The
data required in the SM are prepared with the fuzzy logic
approach. Fuzzy logic provides a new systematic way of
thinking by which complex systems can be understood
without detailed precision but nevertheless accurately with
only a few ordinal parameters.
During the process of model construction, pattern
recognition, and system simulation, many key inputs are
obtained by means of group discussion and consultation.
Fig. 3. Bio-cybernetic criteria of SM.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147136
These open participatory processes focus on communi-
cation, cooperation, and compromise among many different
participants to build consensus for the local development.
These participants include the residents, industries, govern-
ment, environmental groups, and community groups.
Although such a process usually is very time consuming,
it is a remarkable impulse of a new culture of group
learning. Often, many public community meetings are held
and from which, different groups learn to trust, commu-
nicate with, and listen to one another. The process helps
foster a sense of community that is one of the keys to
creating a sustainable community, and a goal of SM as well.
There are nine steps in the operation of SM Tools (see
Fig. 4), they can be divided into three phases: system
definition, pattern recognition, and system simulation and
evaluation. In the first phase, the system components are
compiled according to the characteristics and contents of the
system, which is called a working variable set. The
relationships among the system variables are defined in
the second phase to represent variable behaviors and patterns
for the purpose of system recognition. Finally, with the
variables and their interactions defined, we can perform
scenario simulation and evaluation for a system in the third
phase. The SM applies a top-down approach to deal with
complex system problems. In Fig. 4, the ‘effect system’ is
used to overview the interrelationships among system
variables. The causal relationships of variables are shown
in a system diagram. After the identification of focus issues in
the study area (i.e. traffic, local tourism, etc), ‘partial
scenarios’ are then incorporated to develop simulation
models for each issue, respectively. Each partial scenario
will select from the effect system the variables that are related
to the issue identified, and disaggregate these variables into
more detailed ones. Quantitative relationships between
variables can be identified to simulate the rates of flow.
3.1. Phase 1: system definition
Using the SM tool, a system model is defined by a group
of related variables that cover all important system
components. The process begins with a general system
description of the study area and the identification of
influential factors for local development (see the top of
Fig. 4). These factors are further refined to a limited number
of variables by examining the data available and through
public discussion. This serves as the initial working
variable set.
The completeness of the variable set from a systems
viewpoint is examined with a cybernetic checklist called the
Criteria Matrix. The 18 criteria in the matrix compose
essential parts of a system, including life sectors (7),
physical categories (3), system dynamics (4), and system
relations (4). The variable set should be connected with
these 18 criteria in a well-balanced pattern, which means the
variables construct a complete system. The work of variable
identification and criteria matrix examination will proceed
recursively until a well structured system is defined.
3.2. Phase 2: pattern recognition
Based on the variable set defined in the first phase, the
magnitude of the cause-effect relations between the
variables are examined to identify their functional roles in
the system. There are two key products in the second phase,
the impact matrix and the systemic role of variables. The
impact matrix is an array composed of the effects of each
pair variable. The effect between two variables is evaluated
by the joint discussion of all members of the community for
consensus, referencing the data compiled in advance. The
values of the effect will be summed-up by rows and columns
and we can recognize the different roles of variables with
their influences in the system, as well as characterize the
system behavior accordingly. Each variable can be
categorized into active, passive, critical or buffering
according to the value derived from the impact matrix. A
similar method can be found in sustainability assessment
maps (SAM), which is a graphic tool for displaying
positional information and assisting decision making
(Clayton and Radcliffe, 1996). The result helps planners
to identify the pattern of system behavior and key variables
of the system, which are useful for policy design in the later
phase.Fig. 4. Procedure of SM.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 137
3.3. Phase 3: system simulation and evaluation
The effect system, composed of a causal network of
variables, is organized to serve as a conceptual framework
of system relationships. Both positive and negative causal
networks can be identified from the impact matrix. The
partial scenario of the focus issue can be simulated to
observe system dynamics and the interrelationships between
system variables. In addition, the policy designed to
improve the system can be examined as well. In summary,
the SM tool, a bio-cybernetic evaluation based on the eight
rules above (see Fig. 4) was chosen for its ability to evaluate
the sustainable and long-term viability of the system under
consideration.
4. The case example of Ping-Ding
4.1. Description of Ping-Ding
Ping-Ding is a small village in northern Taipei (Fig. 5). It
is located adjacent to the Yang-Ming-Shan National Park, a
major attraction for residents of Taipei metropolitan during
weekends and leisure time. The population of Ping-Ding
was about 1600 in the year 2000, a figure that has grown
only 2% since 1990.
Due to its specific location and natural resources, Ping-
Ding has become one of the most attractive spots in Yang-
Ming-Shan National Park area. The natural resources there
include several canals that were the major water resource for
both agriculture and daily living in the past. Some very old
religious sites are also interesting. In addition, the
agricultural products such as high mountain vegetables
and orchids are also attractive to the tourists.
The rapid development of tourism has resulted in
positive and negative consequences for Ping-Ding as well
as for Yanh-Mng-Shan National Park. Large numbers of
tourists support the growth of the local economy and
provide the income for residents. The number of orchid
farms and field restaurants has increased to provide service
to tourists. At the same time the development of tourism has
caused some negatives in pollution and turbulence. The
community is facing the conflict between development and
environmental conservation, and how to find a solution to
these problems is a major concern of the residents of Ping-
Ding community currently.
4.2. System definition of Ping-Ding
By reviewing the history and local development of Ping-
Ding, and discussing with participants that included local
residents, the Chi-Shin Conservation Association and the
planning team, we identified some key variables that define
the community. These 26 variables and their definitions are
listed in Table 1. The Variable Set will serve as a basis for
the follow-up procedure. These variables will be refined
within the criteria matrix below.
Following the step of identifying system variables, a
Criteria Matrix is applied to check the system completeness
in relationship to four categories of criteria, which are the
sectors of life, the physical category, the dynamic category,
and the system relationship. Each category contains several
sub-criteria that in total are 18 items. Each of the system
variables was evaluated by the planning team against the 18
criteria, and the matrix was marked with a blank, W, or X to
show not applicable, partly applicable or fully applicable.
Table 2 shows the criteria matrix of Ping-Ding. Take the
variable ‘Canals’ as an example, because it is highly related
Fig. 5. Location of Ping-Ding.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147138
Table 1
System variables and definitions of Ping-Ding
No Variables Definitions
1 Canals Canal is the most important and unique resource of the Ping-Ding community. It characterizes
the community in terms of the function of agriculture irrigation and tourism. In addition, it is
the water resource of daily life of Ping-Ding community
2 Cultural activity The local cultural activity and group activity in front of the temple square has lasted for
hundreds of years as a special event of Ping-Ding. It forms the center of life for the residents of
Ping-Ding and attracts tourists as well
3 Tourism crops Orchids and some high mountain crops are the main tourism crops of Ping-Ding because of the
geographical conditions here. The high mountain vegetable is one of the attractions that bring
people to the community. The restaurant provides tourists with fresh and home raised
vegetables
4 Hiking trails Along with the canal system, there are some hiking trails in the mountain area that provide
access for tourists to reach the beauty of the community. To maintain the trails in good
condition becomes an important issue for the development of the tourism industry of Ping-
Ding
5 Local industry The local industry of Ping-Ding is composed of agriculture, tourism, farms, and restaurants.
They are connected to each other and need more tourists to support them. However, there is a
potential concern that pollution caused by large numbers of tourists could affect the
development of local industry
6 Marketing promotion Since the agriculture product and tourism are the main industries of Ping-Ding, it is important
to make them known to the public. An effective marketing approach will promote the
popularity of Ping-Ding
7 Irrigation resource To support local agriculture and the tourism industry of Ping-Ding, the irrigation system has to
be maintained properly
8 Agriculture Traditional agriculture is facing challenges from both tourism and geographical
conditions. Large numbers of tourists cause pollution. In addition, the farmlands of Ping-
Ding community are subdivided too finely, such that they reduce productivity and
sustainability
9 Competition There are more and more places like Ping-Ding that develop local tourism to attract
people for recreation activity. It is a major issue for Ping-Ding to keep competitive with
other locations in terms of promoting the community image and products
10 Water pollution Water pollution is becoming an issue of greater concern in the community. Several causes of
water pollution are identified including the wastes of tourists, waste water from restaurants,
bio-chemical waste of agriculture, etc
11 Waste treatment There are large amounts of solid wastes and water wastes brought by tourists. To treat waste
costs much money, but it is necessary because the development of the community depends
heavily on environmental quality
12 Recycling Much of the waste of Ping-Ding is reusable, such as leftover food from restaurants
13 Employment To give Ping-Ding an opportunity for self-growth, it is important to keep local employment at
a minimal economic level
14 Culture industry As local culture activity becomes the main attractor of tourists, how to develop the culture
industry becomes a main concern of the Ping-Ding community. The local culture industry of
Ping-Ding includes activity in the temple square, orchid farms, etc
15 Tourist turbulence Large numbers of tourists come to Ping-Ding on the weekends to enjoy the beautiful scenery,
cultural activity and agriculture products. However, they also bring pollution of many types
such as waste, noise, congestion, etc. to the community. It is a concerning issue how to deal
with the turbulence for the local residents
16 Traffic congestion The local network of Ping-Ding is not designed to support the heavy load of tourists in the
weekends. Both tourists and residents suffer from traffic congestion that could reduce the
attraction of the community
17 Local Transportation Appropriate transportation planning such as buses and traffic control can help the traffic
congestion brought by the tourists
18 Accessibility to
water resource
The canal system and Nei-Liao stream compose the water resource of Ping-Ding. It is one of
the main tourism resources that attract people to visit. It is important to promote the
accessibility to these water resources by providing proper facilities
19 Infrastructure To accommodate the large amount of tourists and treat wastes, the infrastructure has to be
planned and developed properly including waste water treatment, solid waste treatment, local
network, etc
20 Community conscience Ping-Ding is a community with small population, isolated from the metropolitan area. The
community conscience is important for local development
(continued on next page)
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 139
to both the criteria of ‘human ecology’ and ‘natural
balance’, we mark both cells with X. It also belongs to
conomy because it brings tourists. Therefore, it gets a W.
Finally, the resulting matrix is quantified and summed
vertically with a 0 for a blank cell, 0.05 for W, and 1 for X,
and the final scores are listed in the bottom row of the
matrix.
The Criteria Matrix serves as a tool to assure that the
Variable Set sufficiently represents the system from a
cybernetics viewpoint. It is an instrument that assists us to
modify the set of variables to address all aspects of the
system. If a zero appears in the vertical sums, or the
numbers show a strange distribution, it means that there
might be some important system components missing and
we have to check the variables again and change their
descriptions. This process could be repeated several times
until we have a sound matrix. In Table 2, the higher scores
of the ‘sector of life’ for Ping-Ding are ‘economy’ (13) and
‘human ecology’ (13), which means that the system
variables are salient along these two criteria and reveals
the importance of these two factors for local development. It
can be seen in Table 2 that the criterion ‘matter’ has the
highest score (12.5) over ‘energy’ and ‘information’ in the
physical category, which tells us that material entities such
as the tourism industry and agriculture are major com-
ponents of Ping-Ding system. In addition to these
characteristics internal to the system, the high scores within
the last three criteria of the category of system relations
(11.5, 13.5 and 9) remind us that Ping-Ding is an open
system and we should take into account the external factors
in the process of policy formulation.
4.3. Pattern recognition of Ping-Ding community
After the system of Ping-Ding community has been
defined, the next step is pattern recognition, in which the
systematic function of each variable of the community will
be identified. This step is based on a pair-wise comparison
of each two variables arranged in an impact matrix as shown
in Table 3. Each cell in the impact matrix reveals the direct
influence of the vertical variable on the horizontal variable,
e.g. the cell in the second column and 4th row shows the
influence of ‘hiking trails’ on ‘cultural activity’. In the
Sensitivity Model, the effect is classified into no signifi-
cance, low significance, medium significance and high
significance, and expressed as 0, 1, 2, and 3 accordingly.
The process of creating the impact matrix involves group
discussion. Three different groups, consisting of residents,
experts, and planning faculty, are requested to discuss and
fill out the matrix separately. After all three groups have
filled out their matrix, the three groups then work together to
create the ‘consensual matrix’. At the same time, the
description of the variables is partly revised and redefined in
such a way that each group can agree on the assessment.
The values in the last two columns and rows of the Impact
Matrix (Table 3) provide us with the information to identify
the role of each variable in the system. When we sum up the
numbers of one row to the right, we get the active sum (AS) of
the corresponding variable. It shows how strongly any
variable effects on the other variables of the system. If a
variable has a relatively high AS, like ‘culture industry (14)’
with 45, any change in that variable will affect the system
significantly, even a small change. In contrast, if the AS of
Table 1 (continued)
No Variables Definitions
21 Environment quality The environmental quality of Ping-Ding community is worse-off because of tourists’
turbulence and waste from tourism activities. It will reduce the attraction of the community
and affect the living of the residents if the environmental quality can not be maintained. How to
keep a balance between the tourism industry and environmental quality becomes more and
more critical for the community
22 Local security There are several security issues of the community. The first is the large amount of illegal
buildings that are disorderly distributed in the area. The second is a shortage of disaster
mitigation resources such as medical assistance. And the third is the disturbance of the tourists
23 Local governance Ping-Ding is composed of a basic political unit, Li. Although the local authority does not have
full administrative power and resources, the consensus of all residents will encourage local
development
24 Education inputs Education investment for the education of environmental concerns and protection, in addition
to the typical education of elementary school, etc. How to construct environmental
consciousness among the residents of Ping-Ding, and facilitate the environmental education of
tourists will be an important force in the development of Ping-Ding
25 Dependence on
imported resource
Since Ping-Ding location is isolated in the Yang-Ming Mountain National Park area, away
from the Taipei Metropolitan area, most matters of daily living are brought from outside areas,
except agricultural products. It is one of the input factors from outside of the system
26 Community image Located around the Yang-Ming Mountain National Park, Ping-Ding has the opportunity to
build its own image as a recreation destination. The components of its image include cultural
activity, the tourism industry and its landscape
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147140
a variable is a small number, this variable has to change
dramatically before it produces a significant effect on the
other variables of the system. When we add the numbers in a
column, we can get the passive sum (PS) of a variable,
showing the extent to which the variable is affected by other
variables in the system. A high PS value such as ‘community
image (26)’ means that as soon as something happens within
the system, this variable will be affected significantly. On the
other hand, a small PS means that within the system, a lot can
happen without changing this variable, e.g. ‘irrigation
resource (7)’.
Since AS and PS represent a one directional effect,
two other indices are useful for describing the role of a
variable in a system. They are P, which represents the
product of AS and PS, and Q, which is the quotient of
AS over PS. A variable with a high quotient value and a
high product value, such as ‘community conscience (20),’
means that it is a dominant variable in the system. A
variable with high quotient value while small product
value, such as ‘accessibility to water resource (18),’
means that it will influence the system with a clear
though weak voice. With the aid of P and Q, we can
interpret the role of the variable of the system more
synthetically. In Fig. 6, each of the variables is located
along the four indices AS, PS, P, and Q, which creates a
field of tension between active, critical, reactive, and
buffering. This provides us with the first strategic
indications by expressing the four indices in a conceptual
way. By their location within this grid, the fields depict
the roles of the variables—an answer produced by the
particular system, but specific for each variable.
According to the above rules, all the variables of the
system are plotted in Fig. 6. We can see that the variables
‘cultural activity (2)’, ‘local industry (5)’, ‘marketing
Table 2
Criteria Matrix of Ping-Ding
Note: Null: not applicable (0); W: partly applicable (0.5); X: fully applicable (1)
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 141
promotion (6)’, ‘culture industry (14)’, ‘community con-
science (20)’, ‘education inputs (24)’, ‘dependence on
imported resource (25)’, and ‘community image (26)’ are
the critical variables in Ping-Ding, which means these
variables are the major driving force of local development.
Some of the variables are worth further investigation.
‘Community image (26)’ is important for Ping-Ding
community because it helps attract tourists. However, its
low Q score 55 indicates that the image of Ping-Ding
community would be affected significantly by many other
variables. This implies that the community and development
policy should pay more attention to preventing negative
impact on the community image. Moreover, the variables
with high values of Q including ‘hiking trails (4)’, ‘waste
treatment (11)’, ‘agriculture (8)’, ‘water pollution (10)’, and
‘tourist turbulence (15)’ would be instrumental as the control
variables of policy design. This information on the roles of
variables within the system can help us formulate policy to
correct or guide the development of Ping-Ding as will be
discussed in Section 4.4 on system simulation.
4.4. System simulation and evaluation
At the beginning of the third phase of SM, a conceptual but
comprehensive causal network of system variables, called
Effect System, is built to illustrate the system framework of
Ping-Ding. The effect system is composed of positive and
negative directional links of variables (see Fig. 7). These
interactions are based on the actual interrelation between
each pair of variables. It can be compared with the values of
the impact matrix to see whether there is a significant
difference between the actual network and consensus
influence, i.e. a 2 or 3 in the impact matrix that does not
connect in the effect system. This could be caused by either a
missing link or an inconsistency between actual and
consensus. We have modified the effect system in this way
repeatedly. Note that since it is too complicated to present all
links of variables, it is more effective to focus on the main
links in the effect system. As a feedback system, there are
over hundreds of feedback loops in which 299 are positive
and 321 are negative. We can examine this more closely by
Table 3
Consensual impact matrix of Ping-Ding
Note: 1. Summation of rows, 2. Product of row sum(AS) and column sum(PS), 3.Summation of columns,
4. Quotient of row sum(AS) and column sum(PS) £ 100.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147142
browsing the list of all feedback loops and any concerning
combination of variables.
4.5. Partial scenario simulation
The effect system gives us a complete but abstract picture of
the internal connections of Ping-Ding. It would be too
complicated to conduct a simulation for the entire system.
Instead, to perform the system simulation in SM, we have to
zoom into a specific part of the system. It is called a partial
scenario (PS). The purpose of a partial scenario is to open the
system and examine the cybernetics of the interesting parts of the
system for the purpose of problem diagnosis, process tracing and
policy testing (Wang, 1994). Since each of the partial scenarios
originates from the main effect system and is based on the main
set of variables, a connection to the total system will always exist
(see Fig. 7). The process of building partial scenarios is begun
with a discussion, either in the form of a one-to-one interviews or
a workshop. In this study, all partial scenarios were sketched
through discussion with local participants, including planners,
local residents and Chi-Sing Conservation Association.
On the basis of system relationships and discussion with
planners, interest groups and local resident representatives, we
identified several issues that were of greatest concern to the
community, including canal reservation, development of the
tourism industry, land utilization, cultural activity and
community environment. Each issue is the focus of a partial
scenario, along with the interlinked variables that were defined
in the previous stages. Though these concerned issues are
presented in partial scenarios individually, they are connected
by means of the connecting variables that exist in more than
one partial scenario. This allows us to perform an integrated
simulation for all the concerned issues and have a compre-
hensive understanding of the development of Ping-Ding.
In addition to the inter-linkage of a scenario, the current
value of each variable and the function of each interlink are
required for simulation. The current value, which represents
the status of each variable and ranges from 1 to 30, is
decided by means of the comprehensive index and
consultations. Take the variable canal as an example; it is
in moderate condition that is not polluted, therefore we
mark it with 18. These values serve as the initial points in
the simulation. In Ping-Ding community, the variables
marketing promotion (6), competition (9), recycling (12),
employment (13), and culture industry (14) are set lower
than 10 which means that these variables are currently in
poor condition. Most variables are in a moderate situation.
Note that the traffic of Ping-Ding community is very
congested, especially during the weekends. The value is set
to 21 with an optimal status set to the lower end.
The function of each interlink in SM is defined such that the
vertical axis is the status value of the source variable and the
horizontal axis is the percentage change of the effected
variable. In Fig. 8, the effect that local security
has on community image is illustrated in the lower left.
When the status value of local security is lower than 3, then
the value of community would diminish by 2%. The value of
community would not be affected when the value of local
Fig. 6. System roles of the variables of Ping-Ding.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 143
security is between 5 and 20. If the value of local security
becomes higher than 20, the value of community image will
increasebya corresponding percentage. The right side ofFig. 8
shows the negative impact that tourist turbulence brings to
local security. These variable relationships are defined by
means of data mining and group discussion.
To obtain a baseline, a 15-year simulation of Ping-Ding
without policy intervention was conducted. The results of
this baseline simulation of some partial scenarios are shown
in Fig. 9. Fig. 9(a) is the simulation of the partial scenario
concerned with the canal resource. It shows that agriculture
and the physical environment facilities of Ping-Ding rise
Fig. 7. Relationship of effect system and partial scenarios.
Fig. 8. Functions of positive and negative interlinks.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147144
steeply and reach the top after periods 7 and 9 years. In that
same time, the quality of water declines. The baseline result
of the partial scenario of local industry is shown in Fig. 9(b).
It shows that the agriculture of Ping-Ding will decline with
the development of tourism crops, and the canal resource
will be affected by the increase of tourism corps and local
industry. Concerning the community living environment,
Fig. 9(c) depicts that over development of the tourism
industry and local industry have significant impacts on local
security and water quality. These baseline simulations
demonstrate possible future problems for Ping-Ding if there
is no effective policy to correct the problems.
For the purpose of solving the possible problems shown
in the baseline simulation above, the community proposes
some possible policies by gathering information from the
previous stages and through discussion with the groups
concerned. Some variables that are critical in the system are
chosen as the control variables. This can be done by
referring to the variables located in the upper-right corner of
Fig. 6. The results of policy simulation are shown in Fig. 10.
These variables including ‘agriculture (3)’, ‘tourism crops
(8)’, and ‘tourist turbulence (15)’ are controlled to a certain
level to avoid over-development. Fig. 10 shows the
simulation policy result in which the upper part is
the simulation result and the lower part is the control of
the policy variable. As shown in Fig. 10(a), the variable
agriculture is controlled to maintain a certain level after
time period 8. This raises the water quality trend above
Fig. 9. Baseline simulation of Ping-Ding.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 145
the declining trend in the baseline simulation. Control of
agriculture can be achieved with the coordination of the
local agricultural association and farmers.
In the second scenario, we considered controlling the
variables ‘tourism crops (8)’ and ‘tourist turbulence (15)’.
Although the development of tourism is important to the
local economy, it could be a potential crisis to community
life without appropriate constraints. Policy that controls
tourism crops and tourist turbulence was simulated and the
result is shown in Fig. 10(b). The upper part of 10(b) is the
simulation result and the bottom of 10(b) shows the control
of the variables. Both variables are controlled from time
period 7 and we reduce control after period 9 to avoid
harming the local economy by over control. Fig. 10(b)
shows that the policy helps keep the variables local security
and water quality from getting worse and remain stable after
period 8. Other partial scenarios are simulated in the same
way to observe how key variables can be controlled to
improve the system.
To evaluate the effect of proposed policy on local
sustainable development, a Bio-cybernetic assessment was
carried out at the end of session three. The evaluation is
based on the eight rules introduced in the second section.
Each criterion is marked with a number ranging from 0 to
100 to show the healthy status of the system before and after
policy control. Table 4 compares evaluation before and after
Fig. 10. Simulations of policy intervention of the partial scenarios.
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147146
policy intervention. The evaluation results, in which five out
of eight criteria get higher scores, show that the overall
system is improved with policy intervention. However, the
score is evaluated worse under the criterion of the jiu-jitsu
principle because the policy intervention uses direct control,
not the market mechanism. This result provides information
for the next step of policy formulation.
5. Discussion and concluding remarks
Systems thinking about the development of sustainable
communities is applied in this paper in the analysis of the
Ping-Ding community by means of the SM. The application
of SM to community development for sustainability in
Taiwan was a very positive experience for us. The tool
raises the accessibility of local residents and interest groups
to development policy for their community. It brings the
people together who are concerned with local development
and provides them with a convenient tool to share their view
and test the possible policy outcomes. Traditionally, there
have been fewer meetings, and they lack concrete and
operational methods for the participants to have clear
images of a what-if scenario.
SM provides the planner with a convenient and effective
tool for the process of public participation and consensus, a
key element of the implementation of sustainable develop-
ment. It provides planners with information about the roles
of system variables for the purpose of policy formulation. In
the case of Ping-Ding community, the simulation of partial
scenarios suggests that the tourism industry of the commu-
nity should be carefully controlled to avoid the negative
impact caused by tourist turbulence and pollution.
This semi-quantitative approach provides a practical
solution for dealing with the complicated relations among
variables. However, the specification of the interlink
function for each pair of variables requires a great deal of
consultancy work. Also, the interpretation of the eight Bio-
cybernetic rules of system evaluation depends heavily on
the experience and subjectivity of the planners. These issues
may hinder adoption of the SM model by Taiwan’s planning
environment and several follow-up studies toward reducing
operation costs are in the process to improve the use in
planning practice.
References
Beatley, T., Manning, K., 1997. The Ecology of Place, Island Press,
Washington, DC.
Clayton, A., Radcliffe, N., 1996. Sustainability, A Systems Approach,
Earthscan Publications, London.
Forman, R., 1995. Land Mosaic: the Ecology of Landscapes and Regions,
Cambridge University Press, Cambridge.
Hodge, R.A., Hardi, P., 1997. The Need for Guilelines: the Rationale
Underlying the Bellagio Principles for Assessment. In: Hardi, P., Zdan,
T. (Eds.), Assessing Sustainable Development: Principles in Practice,
The International Institute for Sustainable Development, Winnipeg, pp.
7–20.
Huang, S.-L., Wong, J.-H., Chen, T.-C., 1998. A framework of indicator
system for measuring Taipei’s urban sustainability. Landscape and
Urban Planning 42, 15–27.
Huang, S.-L., Chen, C.W., 1999. A system dynamics approach to the
simulation of urban sustainability. In: Brebbia, C.A., Uso, J.L.
(Eds.), Ecosystems and Sustainable Development II, WIT Press,
Southampton.
Jorgensen, S., 1997. Integration of Ecosystem Theories: A Pattern, second
ed, Kluwer Academic Publisher, London.
Kildow, J.T., 1992. The earth summit: we need more than a message.
Environmental Science and Technology 26(6), 1077–1078.
Lachman, B., 1997. Linking Sustainable Community Activities to Pollution
Prevention: A Sourcebook (Santa Monica, RAND).
Maureen, H., 1999. The Guide to Sustainable Community Indicators,
second ed, Sustainable Measures (North Andover, MA).
Rothman, D.S., Robinson, J.B., Bigg, D., 2002. Signs of Life: Linking
Indicators and Models in the Context of QUEST. In: Abaza, H.,
Baranzini, A. (Eds.), Implementing Sustainable Development—Inte-
grated Assessment and Participatory Decision-Making Processes,
Edward Elgar, Cheltenham, UK.
Rotmans, J., van Asset, M., 2002. Integrated Assessment: current practices
and challenges for the future. In: Abaza, H., Baranzini, A. (Eds.),
Implementing Sustainable Development—Integrated Assessment and
Participatory Decision-Making Processes, Edward Elgar, Cheltenham,
UK.
Vester, F., 1988. The bio-cybernetic approach as a basis for planning our
environment. Systems Practice 1(4), 10–16.
Vester F., 1999. Die Kunst vernetzt zu denken (The Art of Interlinked
Thinking). (Deutsche Verlagsanstalt, Stuttgart).
Vester F., Hesler A., 1982. Sensitivity Model. (Frankfurt/Main, Umland-
verband Frankfurt).
Wang, R., 1994. Planning the Ecological Order—A Human Ecological
Approach to Urban Sustainable Development. In: Wang, R., Lu, Y.
(Eds.), Urban Ecological Development: Research and Application,
China Environmental Science Press, Tianjin.
WCED, 1987. Our Common Future, Oxford University Press, Oxford.
Zachary, J., 1995. Sustainability community Indicators: guideposts for
local planning, Community Environmental Council, Santa Barbara.
Table 4
System evaluation of Ping-Ding
Bio-cybernetic criteria Evaluation
Negative feedbacks must dominate positive feedbacks ¼
Vitality of the system must be independent
from quantitative growth
¼
System must work function-oriented
and not product-oriented
"
Use of existing forces (principle of jiu-jitsu)
instead of fighting (boxing-method)
#
Multiple usage of products, functions
and organizational structures
"
Recycling processes for the utilization of waste and sewage "
Symbiosis reciprocal usage of difference by
the employ of coupling and exchange, and
"
Biological design of products, procedures and
forms of organizations by feedback planning
"
S.-L. Chan, S.-L. Huang / Journal of Environmental Management 72 (2004) 133–147 147